Finance Agent v2 - Precedents
Professional Work · 2026-05-19
The Precedents split of Finance Agent v2. This child benchmark separates a source-reported subtask or subtrack from the parent aggregate so scores at different grains do not share one benchmark_slug.
Top models (higher is better)
| Model | Score |
|---|---|
| Gemini 4 Argon | 49.8 |
| Muse Spark Open | 40.8 |
| Gemini 3.5 Flash | 36.4 |
| Gemini 3.8 Flash | 36.1 |
| Muse Spark 1.2 | 35.4 |
| Muse Spark 1.3 | 35.2 |
| Muse Wonder Full | 35.0 |
| Gemini 3.7 Flash | 33.8 |
| Muse Spark 1.1 | 33.8 |
| Claude Fable 5.1 | 32.2 |
| Muse Wonder | 32.1 |
| MiMo-V2.6-Pro | 30.9 |
| Gemini 3.6 Flash | 30.8 |
| Hy4 Preview | 30.6 |
| GPT-6 Astra | 30.3 |
| MiMo-V2.6-Flash | 30.3 |
| Qwen3.8 Max (0902) | 29.5 |
| Ember-1 | 29.2 |
| GPT-5.6 Luna | 29.1 |
| Sonnet 5 | 28.8 |
| GLM-5.3 Flash | 28.6 |
| GPT-6 Sol | 28.5 |
| DeepSeek V4.1 Flash | 28.2 |
| GPT-6 Luna | 27.4 |
| Claude Haiku 5.5 | 27.0 |
| DeepSeek V4.1 Flash 1017 | 26.5 |
| GPT-5.6 Sol | 26.2 |
| Mistral Large 4 Preview | 26.2 |
| Gemini 3.5 Flash-Lite | 26.0 |
| Grok 4.7 | 25.9 |
| GPT-6.1 Sol | 25.8 |
| Opus 4.8 | 25.6 |
| GLM-5.3 | 25.1 |
| GPT-5.6 Terra | 24.8 |
| Inkling | 24.8 |
| DeepSeek-V4-Flash-0731 | 24.1 |
| Kimi K3 | 23.6 |
| MiniMax M3 | 23.2 |
| GPT-5.4 Mini | 22.1 |
| Grok 4.6 | 21.9 |